Serum ammonia is a significant biochemical marker for the diagnosis of hepatic dysfunction, urea cycle disorders and other metabolic disorders [1]. The main sources of ammonia production are amino acid metabolism, activity of intestinal bacteria and degradation of nitrogenous compounds [2]. The liver regulates levels of ammonia in the blood by catalyzing its conversion to urea via the urea cycle which is normally present in the liver, and under physiological conditions, it is harmlessly excreted from the body [3].

High blood ammonia levels may be due to a reduced capacity of the liver to detoxify ammonia or to an imbalance of ammonia metabolism, especially in people with liver disease and some inherited metabolic disorders [4]. Serum ammonia should therefore be measured accurately in order to be interpretable in the laboratory and provide reliable clinical assessment.

Preanalytical factors such as specimen collection, transit, processing, and storage are especially critical for the measurement of ammonia [5]. Ongoing metabolism of the blood cells after separation may cause a rise in ammonia concentration, particularly if serum or plasma separation is delayed, or if there is a release of ammonia-containing components after separation [6]. The temperature has been identified as a critical factor affecting these processes and the stability of ammonia measurements can be affected by storage conditions [7]. Therefore, it is recommended to process the specimens as soon as possible and to keep the appropriate temperature to reduce the preanalytical variation [8]. Variability in laboratory procedures could be a factor in the variability of results, which could make interpretation of ammonia levels more difficult between different clinical settings.

Other factors such as light exposure, temperature, and storage time of the specimens can also affect the accuracy of lab measurements [9,10]. Though it is known that temperature and processing time are factors to consider in ammonia testing, it is important to carefully evaluate the effects of light exposure as well as the interaction between temperature and processing time in defined laboratory conditions [11,12].

 These factors may be investigated to help identify changes in the measured ammonia concentration and to differentiate real biochemical changes from those occurring after samples were collected [13,14]. It is therefore more appropriate to consider the standardized assessment of light exposure, storage temperature and time to understand the stability of the serum ammonia and the potential limitations of routine sample handling.

Research Objective

To determine the effects of light exposure, temperature, and duration of storage on measured serum ammonia levels under controlled laboratory conditions.

Study Design and Experimental Framework In a laboratory-based experimental study, the influence of light exposure, temperature and storage time on measured serum ammonia was explored. A 2x2x6 repeated-measures factorial design was suggested with two light conditions, two storage temperatures and six time points for measurements. Aliquots were taken from each independent specimen and divided to the different experimental conditions to reduce inter-individual biological variability and to assess the individual as well as combined effects of the three factors. Study Setting and Duration The study was carried out in AFIP Rawalpindi, Pakistan, at clinical biochemistry laboratory of the hospital for two years period (August 2022 to August 2024). All specimens were to be prepared under standardized laboratory conditions, stored in a standardized manner, and all biochemical analyses were to be performed with calibrated equipment and standardized procedures. In the final manuscript, the name of the hospital and the laboratory should be given. Sample Size Determination and Statistical Power Sample-size planning was based on a repeated-measures factorial design, with serum ammonia concentration as the primary outcome. Preliminary assumptions included a medium effect size (Cohen’s f=0.25f = 0.25f=0.25), a significance level of 0.05, statistical power of 80%, and an assumed within-specimen correlation of 0.50. A target of 30 evaluable specimens was proposed, with recruitment of up to 38 specimens to allow for approximately 20% specimen loss or exclusion. These values represent preliminary planning assumptions rather than a confirmed power calculation. The final sample size should be justified using pilot data and a power analysis that considers the main effects and interactions among light exposure, temperature, and storage time. Study Population and Specimen Selection The samples of serum were to be collected from eligible women visiting the selected health care unit in Pakistan. The plan to be followed was to collect venous blood samples from adults aged 18-60 years in a prospective manner, if approved by the institutions and after obtaining informed consent from the participants. One independent specimen was to be contributed by each participant, to be divided into aliquots to be subjected to the experimental conditions. If laboratory specimens remain then the study design and description of the sources of specimens need to be changed. Inclusion and Exclusion Criteria Eligible specimens were those that were well labeled with enough samples for baseline and further analysis. The inclusion criteria for the proposed study were specimens collected from adults 18–60 years of age and specimens processed following the study protocol. Haemolytic specimens and those that were too small, wrongly identified or for which collection and processing times were not documented were to be excluded. The final eligibility criterion ought to be in line with the approved protocol and the specimens submitted to be studied. Ethical Approval and Specimen Consent Before participant recruitment or specimen collection, approval was obtained from the institution's research ethics committee (IRB) Blood Collection and Serum Preparation Standardized procedure was used in an appropriate serum collection tube, to collect venous blood. Samples were allowed to clot and were then centrifuged using the laboratory's validated SOP. Blood cells were separated immediately and serum was subdivided into labeled aliquots so as to reduce the number of times cells were handled and risked becoming contaminated. The following were to be recorded: collection-to-separation time, centrifugation speed and time, aliquot volume, and baseline ammonia concentration. The same method of preparing the specimens was used for all the experimental groups. Experimental Conditions and Factorial Study Design Aliquots of the serum samples of each individual specimen were placed into various combinations of light exposure and storage temperature. The level of ammonia was measured at 0, 1, 2, 4, 8 and 24 hours under the two light conditions (light exposure at 1,000 lux, light protection) and under the two temperature conditions (refrigerated storage (4 ± 1°C), room-temperature storage (25 ± 2°C)). In this setup, 24 light–temperature–time combinations could be generated for each specimen when all six of the time points were measured for each combination of light and temperature. Aliquots were separated at each time point if possible. To reduce procedural variation, a standard procedure for aliquot allocation and handling was developed. Light Exposure Conditions For the light-exposed condition, specimens were illuminated with white LED lights at 1,000 lux (at specimen level). Specimens for light protected condition were wrapped in aluminium foil to keep them from light exposure. A calibrated lux meter was to be used to measure light intensity, if available. Light source, distance from the specimens, exposure time and intensity of the illumination were recorded. Aliquots were kept for the same period of time at the same temperature and evaluated at the same time to separate the effect of light. Temperature-Controlled Storage Conditions Aliquots were kept at 4 ± 1°C (refrigerated) and at 25 ± 2°C (controlled room temperature). A calibrated digital thermometer or temperature data logger was used to record temperature, at least every 30 minutes. Specimens were kept at their respective temperatures until the time of the measurement. Data were documented for any deviations in temperature and were taken into account when interpreting the data. Specimen Storage Duration and Sampling Time Points Serum ammonia was measured at baseline (0 hours) and at 1, 2, 4, 8 and 24 hours. Baseline measurements were taken right after the serum was prepared and initial processed. All subsequent measurements were to be made with respect to the established "baseline time. Aliquots were divided into individual time points to reduce repeated opening, handling and temperature fluctuations. The time interval between separation of the serum, storage and examination was actually recorded. Serum Ammonia Measurement and Analytical Procedures Validated clinical chemistry assay method was used to quantify serum ammonia levels, following manufacturer's instructions and laboratory's standard operating procedures. The method proposed was an enzymatic UV based method and the results obtained in the proposed method were expressed in µmol/L using an automated clinical chemistry analyzer. All of the following must be indicated: analyzer manufacturer and model, reagent manufacturer, principle of the assay, calibration procedure, analytical measuring range and reporting units from actual laboratory records. Wherever possible, baseline and subsequent measurements were done with the same analytical technique and instrument. Quality Assurance and Analytical Quality Control Internal quality controls, as recommended by the manufacturer, were used to monitor the analytical quality, employing appropriate internal quality control materials at normal and elevated ammonia levels. Documentation of instrument calibration, reagent validity and quality-control acceptance criteria were documented. The laboratory's analytical validation records or manufacturer's documentation should provide within-run and between-run coefficients of variation (CVs). Analytical errors and quality control problems were checked prior to inclusion in the final measurement. Assessment of Preanalytical Variability and Specimen Stability Specimen stability was assessed by comparing ammonia concentrations across the defined light, temperature, and storage-time conditions against baseline measurements. Absolute changes were calculated as the difference between each subsequent measurement and baseline. Percentage change from baseline was calculated as: % change= Follow-up Value−Baseline value/ Baseline value ×100 Observed changes were interpreted in relation to analytical imprecision and the performance characteristics of the assay. This approach was intended to distinguish storage-associated changes from ordinary measurement variability. Primary and Secondary Outcome Measures The main objective was the difference in the ammonia concentration measured in the serum from the baseline condition in the defined experimental setting. Secondary outcomes were percentage change from baseline and the main and interaction effects of light exposure, storage temperature and storage time. The concentration of ammonia was reported as µmol/L and the same units were applied during the analysis. Statistical Analysis Appropriate statistical software was to be used to analyse the data. Data for continuous variables were summarized as means ± SD or medians and IQRs when distributions were non-normal. The effects of light, temperature, and time with two-way and three-way interactions were planned to be evaluated using repeated measures factorial analysis of variance or a linear mixed-effects model. The model took into account repeated measurements on the same specimen. Model assumptions and residual distributions were evaluated and corrections were made as appropriate. A suitable procedure was used to adjust for multiple comparisons. The results were presented as effect estimates and were the 95% confidence intervals with the corresponding p-values. P < 0.05 was accepted as statistically significant. Data Management and Reproducibility Data were entered into a standardized data structure with coded data identifiers and standardized definitions of variables. The records contained the concentration of ammonia at baseline, experimental conditions, storage time, temperature measurements, assay results, and observations of quality control. Data was verified for completeness, consistency and outliers prior to analysis. The methods used and deviations from experiments were documented, as were the statistical methods, in order to ensure transparency and reproducibility.

Table 1 presents the baseline characteristics of 30 specimens. The mean serum ammonia concentration was 28.64 ± 6.83 µmol/L, while the median was 27.90 µmol/L (IQR: 23.80–32.60), with values ranging from 16.40 to 43.70 µmol/L. The experiment included two light conditions, two storage temperatures, and six measurement time points.

Table 1. Baseline characteristics and serum ammonia concentration

Table 2 shows that serum ammonia concentrations generally decreased slightly under refrigeration but increased at room temperature over 24 hours. At 24 hours, mean concentrations were 27.10 µmol/L (light-protected) and 26.81 µmol/L (light-exposed) at 4°C, compared with 38.25 and 39.70 µmol/L, respectively, at 25°C.

Table 2. Serum ammonia concentrations under different light and temperature conditions over time

Table 3 summarizes absolute and percentage changes from baseline. After 24 hours, ammonia decreased by 1.54 µmol/L (5.38%) under refrigerated, light-protected conditions and by 1.83 µmol/L (6.39%) under refrigerated, light-exposed conditions. In contrast, concentrations increased by 9.61 µmol/L (33.55%) at 25°C with light protection and by 11.06 µmol/L (38.62%) with light exposure.

Table 3. Absolute and percentage changes in serum ammonia from baseline

Table 4 presents the factorial analysis. Storage temperature (+8.45 µmol/L, 95% CI: 7.80–9.10, p < 0.001) and storage time (+0.38 µmol/L/hour, 95% CI: 0.32–0.44, p < 0.001) had statistically significant effects. The temperature-by-time interaction was also significant (p < 0.001), whereas the overall light effect (p = 0.142) and three-way interaction (p = 0.620) were not statistically significant.

Table 4. Factorial analysis of the effects of light, temperature, and storage time

Footnote: Effect estimates represent mean differences or slopes derived from the fitted linear mixed-effects model relative to reference categories (Light-protected, 4°C).

Table 5 presents pairwise comparisons at 24 hours. Light exposure was not significantly associated with ammonia differences at 4°C (−0.29 µmol/L, p = 0.650), but the difference was significant at 25°C (+1.45 µmol/L, p = 0.002). Room-temperature samples also had significantly higher concentrations than refrigerated samples, with differences of +11.15 µmol/L for protected samples and +12.89 µmol/L for exposed samples (both p < 0.001).

Table 5. Pairwise comparisons between experimental conditions at 24 hours

Footnote: Calculations based on estimated marginal means at the 24-hour time point, adjusted using Holm's procedure for multiple comparisons.

In 30 samples, the effect of light, storage temperature and storage time on the measured serum ammonia concentration was investigated. The baseline mean was 28.64 ± 6.83 µmol/L, with a median of 27.90 µmol/L (IQR: 23.80–32.60). The results presented indicate that the separation temperature and storage time played an important role in the separation results. This is in line with the findings by Nikolac et al., who examined the ammonia changes at refrigerated storage for 20 healthy volunteers with only a few preanalytical variables selected [15]. The results highlight the need for uniform specimen collection procedures to use when analyzing ammonia. The most significant factor in the results provided was temperature. The mean ammonia levels of light-protected and light-exposed refrigerated specimens were 27.10 and 26.81 µmol/L after 24 hours. Corresponding values at 25°C were 38.25 and 39.70 µmol/L. The reported temperature effect was +8.45 µmol/L (95% CI 7.80,9.107.80, 9.10, p < .001). Similarly, Goldstein et al. (2017) reported linear increases in measured ammonia during storage that were carried out at 4°C and −14°C, and that were stable for up to three weeks when performed at −70°C [16]. Their study, while not conducted in serum, confirms that the storage conditions and matrix should be taken into account when assessing the stability of ammonia in the plasma. Significant changes were also related to storage duration. A increase of 33.55% and 38.62% were recorded for ammonia after 24 hours in refrigerated and light-protected and room-temperature and light-exposed specimens, respectively. The reported time effect was +0.38 µmol/L per hour (95% CI 0.32,0.440.32, 0.44, p < .001), and the temperature-by-time interaction was +0.35 µmol/L per hour (95% CI 0.28,0.420.28, 0.42, p < .001). Gifford et al., 2018, noted that the level of ammonia increased in stored plasma and that the rate of increase was different between the untreated plasma based on enzyme activity and storage conditions [17]. Their results confirm the importance of dealing with elapsed time and temperature as two interrelated handling variables. The overall effect of light exposure was smaller than the effect of temperature. At 24 hours, the exposed-versus-protected difference was −0.29 µmol/L at 4°C (adjusted p = .650) and +1.45 µmol/L at 25°C (adjusted p = .002). The overall light effect was not statistically significant (+0.42 µmol/L, 95% CI −0.15,0.99−0.15, 0.99, p = .142). In the case of ammonia, the stability and transport, and the centrifugation temperature, are of particular importance in preanalytical handling that were examined by Favresse et al. [18]. But these results are not proof of a direct impact of light on exposure. The nonsignificant three-way interaction (p = .620) in the present results does not show that there is an interaction of light and temperature and time, nor does it warrant further investigation. Real-world laboratory evidence should also be taken into consideration when interpreting the findings. In their hospital setting, Imbert-Bismut et al. did not observe any effect on ammonia results from pneumatic transport and set a waiting period, whereas the storage at −80°C improved the stability [19]. Ingestion of the same by Mercer-Smith et al. (2025) demonstrated that ammonia was stable in samples collected from patients with hyperammonaemia and they also questioned the need for transport on ice in this specific context [20]. These studies have shown that acceptable handling procedures will vary according to the type of specimen, the delay in processing, and the procedures for the laboratory. Thus, the results of the present study cannot be extrapolated to all laboratories or clinical groups. In general, the provided data indicate that for the measured serum ammonia concentrations, room temperature and more extended storage period had a positive relationship, whereas the other light effect was not statistically significant. However, the numerical results and statistical estimates should be tested using the original data on the individual specimens before they are published. These patterns should be validated in further studies with independently replicated specimens, assay performance validated, and predefined stability limits. The results of this experiment may not be directly applied to whole blood or plasma. The findings, if they are confirmed, could assist laboratories in improving their specimen handling processing and decrease the analytical variation that occurs during storage. Strengths and Limitations: This study has a strong advantage in that it follows a factorial experimental design and uses 30 independent samples of serum ammonia in evaluating the effect of light exposure, storage temperature and storage duration on serum ammonia concentrations separately and in combination. Because multiple measurements can be taken over 24 hours and the experimental conditions are standardized, changes over time can be evaluated. This study is, however, limited in a number of ways; sample size was relatively small, the study was conducted in one laboratory, and there was no confirmed independent replication. Additionally, the results provided must be checked against the original data, especially the mixed-effects model results and pairwise comparisons. Moreover, it is unclear if the results from separated serum samples would apply to plasma or whole blood samples, and it is not known if the changes observed would impact clinical decision making. Such restrictions must be taken into consideration prior to reaching any final determinations.

Serum ammonia concentration seems to be affected by storage time and temperature. Ammonia levels increased with time in room-temperature storage conditions and were more stable in refrigerated storage conditions. The overall light effect was comparatively small, but might be different depending on various storage conditions. The results emphasize the need for uniform specimen handling and prompt analysis to reduce the preanalytical sources of variation. These observations require further studies with accurate experimental data and larger sample size to validate the observation and provide suitable recommendations for laboratory handling.

1. B Adeva MM, Souto G, Blanco N, Donapetry C. Ammonium metabolism in humans. Metabolism. 2012;61(11):1495–1511. https://doi.org/10.1016/j.metabol.2012.07.007
2. Gifford JL, Nguyen WNT, de Koning L, Seiden-Long I. Stabilizing specimens for routine ammonia testing in the clinical laboratory. Clin Chim Acta. 2018;478:37–43. https://doi.org/10.1016/j.cca.2017.12.022
3. Gallego-Durán R, Hadjihambi A, Ampuero J, et al. Ammonia-induced stress response in liver disease progression and hepatic encephalopathy. Nat Rev Gastroenterol Hepatol. 2024;21:774–791. https://doi.org/10.1038/s41575-024-00970-9
4. Sepehrinezhad A, Moghaddam NG, Shayan N, Sahab Negah S. Correlation of ammonia and blood laboratory parameters with hepatic encephalopathy: A systematic review and meta-analysis. PLoS One. 2024;19(9):e0307899. https://doi.org/10.1371/journal.pone.0307899
5. da Fonseca-Wollheim F. Preanalytical increase of ammonia in blood specimens from healthy subjects. Clin Chem. 1990;36(8 Pt 1):1483–1487. https://doi.org/10.1093/clinchem/36.8.1483
6. Howanitz JH, Howanitz PJ, Skrodzki CA, Iwanski JA. Influences of specimen processing and storage conditions on results for plasma ammonia. Clin Chem. 1984;30(6):906–908. https://pubmed.ncbi.nlm.nih.gov/6723050/
7. Nikolac N, Omazic J, Simundic AM. The evidence based practice for optimal sample quality for ammonia measurement. Clin Biochem. 2014;47(12):991–995. https://doi.org/10.1016/j.clinbiochem.2014.05.068
8. Korzun WJ. Blood ammonia stability revisited. Clin Lab Sci. 2015;28(3):173–178. https://clsjournal.ascls.org/content/28/3/173
9. Mercer-Smith GW, Appleton M, Hanon ÉA, Bowron A. Blood samples for ammonia analysis do not requiretransport to the laboratory on ice: a study of ammonia stability and cause of in vitro ammonia increase in samples from patients with hyperammonaemia. Clin Chem Lab Med. 2025;63(6):1132–1138. https://doi.org/10.1515/cclm-2024-1304
10. Hedayati M, et al. The impact of pre-analytical variations on biochemical analytes stability: A systematic review. J Clin Lab Anal. 2020;34(12):e23551. https://doi.org/10.1002/jcla.23551
11. Choy KW, Wijeratne N, Doery JCG. What is the most suitable specimen for ammonia measurement? Pathology. 2016;48(Suppl 1):S87. https://doi.org/10.1016/j.pathol.2015.12.240
12. Transportation and handling of blood samples prior to ammonia measurement in the real life of a large university hospital. Clin Chim Acta. 2020. https://pubmed.ncbi.nlm.nih.gov/32735981/
13. Kazi, A. K., Jatoi, A., & Hashmani, H. (2025). An Empirical Evaluation of Technology Acceptance Model in Mobile Devices in Healthcare Industry. Journal of

14. Development and Social Sciences, 6(3), 707–714. https://doi.org/10.47205/jdss.2025(6-III)60
15. Jatoi, A., Bibi, M., Hashmani, H., & Ramish, M. S. (2025). Awareness of Patients’ Rights and the Respect and Dignity Experienced at Hospitals in Karachi, Pakistan. International Journal of Trends and Innovations in Business & Social Sciences, 3(3), 232–243. https://doi.org/10.5281/zenodo.17359794
16. Nikolac, N., Omazic, J., & Šimundić, A.-M. (2014). The evidence based practice for optimal sample quality for ammonia measurement. Clinical Biochemistry, 47(12), 991–995. https://doi.org/10.1016/j.clinbiochem.2014.05.068
17. Goldstein, B. N., Wesler, J., Nowacki, A. S., Reineks, E., & Natowicz, M. R. (2017). Investigations of blood ammonia analysis: Test matrices, storage, and stability. Clinical Biochemistry, 50(9), 537–539. https://doi.org/10.1016/j.clinbiochem.2017.01.002
18. Gifford, J. L., Nguyen, W. N. T., de Koning, L., & Seiden-Long, I. (2018). Stabilizing specimens for routine ammonia testing in the clinical laboratory. Clinica Chimica Acta, 478, 37–43. https://doi.org/10.1016/j.cca.2017.12.022
19. Favresse, J., Despas, N., Boland, L., Gruson, D., Laterre, P.-F., Nassogne, M.-C., Vincent, M.-F., & Fillee, C. (2018). Preanalytics of ammonia: Stability, transport and temperature of centrifugation. Clinical Chemistry and Laboratory Medicine, 56(3), e65–e68. https://doi.org/10.1515/cclm-2017-0751
20. Imbert-Bismut, F., Payet, P.-E., Alfaisal, J., Munteanu, M., Rudler, M., Sultanik, P., Alkouri, R., Sakka, M., Djavoudine, S., Dever, S., Mestari, F., Bonnefont-Rousselot, D., Poynard, T., & Thabut, D. (2020). Transportation and handling of blood samples prior to ammonia measurement in the real life of a large university hospital. Clinica Chimica Acta, 510, 522–530. https://doi.org/10.1016/j.cca.2020.07.048
21. Mercer-Smith, G. W., Appleton, M., Hanon, É. A., & Bowron, A. (2025). Blood samples for ammonia analysis do not require transport to the laboratory on ice: A study of ammonia stability and cause of in vitro ammonia increase in samples from patients with hyperammonaemia. Clinical Chemistry and Laboratory Medicine, 63(6), 1132–1138. https://doi.org/10.1515/cclm-2024-1304